WO2001098039A1

WO2001098039A1 - Parallel manipulator provided with a truss structure

Info

Publication number: WO2001098039A1
Application number: PCT/DK2001/000433
Authority: WO
Inventors: Jakob Steinicke; Ole Graa Jakobsen; Anders Stengaard SØRENSEN; Lars Overgaard
Original assignee: Meganic Aps
Priority date: 2000-06-20
Filing date: 2001-06-20
Publication date: 2001-12-27
Also published as: AU2001265832A1; EP1296805A1

Abstract

A device for movement and positioning of an object in space is provided which has a parallel manipulator structure comprising a number of rigid force transferring elements defining a truss. The device is modular and a large range of kinematics characteristics may be obtained by combining numbers of the parallel manipulator structures differently. The device has actuators and a control system for controlling the movement and orientation of an object in space.

Description

PARALLEL MANIPULATOR PROVIDED WITH A TRUSS STRUCTURE

Field of the Invention

The present invention relates to a device for movement and positioning of an object in space. The device according to the invention may in particular be a robot. The robot is made up of a plurality of devices. The device and the robot according to the invention are in particular useful for performing automated assembling, welding, painting, deburring etc. It can also be applied in entertainment applications such as animatronics, motion simulator platforms etc.

The present invention is primarily concerned with the kinematics structure of manipulators, in particular those incorporating so-called parallel mechanisms, as well as with the control of such manipulators and mechanisms.

Description of the Prior Art

Generally speaking, two types of mechanisms for manipulators and robots have been provided by the prior art, namely serial mechanisms and parallel mechanisms.

It has been found that serial mechanisms have limited reach-ability and load capacity while having relatively poor stiffness-to-weight and load-to-weight ratios. On the other hand, serial mechanisms have the benefit of large dexterity. Parallel mechanisms have been found to have stiffness-to-weight and load-to-weight ratios which are superior to the corresponding ratios of serial mechanisms but the dexterity is limited.

US 5,699,695 discloses a spatial, parallel-architecture robotic wrist. The wrist has a base plate and a distal plated coupled together by three symmetric, parallel chains of links, wherein each of the chains consists of five revolute joints in series, the first two being perpendicular and having intersecting axes, the second through forth intersecting a point, and the fourth and fifth again perpendicular and having intersecting axes. According to the description of US 5,699,695, the wrist disclosed therein provides a closed form solution, or a solution based on analytical expressions, for solving the kinematics of the wrist. The wrist has six links arranged in three chains of two links each and with three actuators being mounted in fixed relationship to the base plate. The three actuators cause the three links connected to the base plate to each pivot with a single degree-of-freedom relative to the base plate. Each of the three chains of links includes a three degree-of-freedom joint between its two links. The distal plate is connected to three of the links with the connection between the distal plate and each of the three links allowing a single degree- of-freedom between the distal plate and each of the three links.

US 4,806,068 relates to a robot manipulator having co-operating linear and rotating actuators placed on a base plate. According to the description of US 4,806,068, the mechanism of the robot manipulator may be utilised for controlled movement of a tool.

Description of the Invention

An object of the present invention is to provide improvements to known manipulators in respect of reach-ability, dexterity as well as stiffness-to-weight and load-to-weight ratios.

Accordingly, in a first aspect the invention provides a device for movement and positioning of an object in space, said device comprising:

- a proximal member and a distal member facing each other and being interconnected by n>2 linking members, each linking member comprising a first and a second linking member,

- power driven means for causing relative movement between the proximal member and the first linking members,

- each of the first linking members having a first rotational joint for connection of the first linking member to the proximal member and a spherical joint for connection of the first linking member to the second linking member,

- each of the second linking members having a first rotational joint for connection of the second linking member to the distal member and a second spherical joint for connection of the second linking member to the first linking member, the first and second linking members thereby being spherically interconnected by their respective second joints,

the first linking members comprising: - at least two rigid force transferring elements, each of the elements at a first end being mutually interconnected in a connection point and at a second opposite end being connected to the first linking member so as to define a truss, and

- a third joint for connection of said linking member to the power driven means, the joint being arranged substantially in the connection point.

Accordingly, a construction is achieved wherein substantially no torque is being excelled on the second linking member when a force is excerted thereupon by the first linking member and/or the distal member. Instead tension or compression is applied at the ends of the second linking members. Accordingly, the second linking member may be build with smaller dimensions than linking members wherein torque is excerted e.g. the first linking member. This results in a lower weight of the second linking member. At the same time no or little torque is applied to the two force transferring elements connected to the first linking member, resulting in similar low weight of the two force transferring elements.

The lower weight confers a smaller deflection caused by the weight of the parts of the device, thereby allowing for more precise control of the object connected to the n'th module. Consequently, more modules may be comprised in the manipulator, resulting in higher reachability, dexterity and flexibility.

The two force transferring elements may be connected to the first linking members in the vicinity of the first and second rotational joints of the first linking members. As result no or little torque is applied to the first linking member. Thereby the dimensions of the first linking member can be smaller resulting in a lower weight. In this embodiment no or little forces are applied to any of the parts of the device resulting in lower weight of the device as a whole.

In preferred embodiment of the invention, the load-to-weight ratio of the device is preferably at least 0.4, such as at least 0.6 or preferably at least 0.7 or at least 0.8 or higher, such as at least 0.9 or 0.95. The load-to-weight ratio is defined the ratio between the load which the device or a part of the device may carry and the weight of the manipulator or the module, respectively.

More than two force transferring elements may be interconnected to the first linking members, thus forming space truss structures. As an example the space truss structures may consist of at least 6 force transferring elements connected in at least 4 connection points. The first joints of the first linking members may allow rotation around pivot axes parallel to lines extending between a first and a second of the 4 connection points. The second joints of the first linking members being disposed in the vicinity of a third of the 4 connection points. The third joints of the first linking members being disposed in the vicinity of a fourth of the 4 connection points. As an example the 6 force transferring elements may have a length so that the space truss structure of the first linking members defines a tetrahedron.

An alternative embodiment of connecting the force transferring elements to the third connection points and thereby to the first linking members may be to attach moment arms to the first linking members. As an example this could be done so that a first end of a moment arm is connected to the first linking member and a second end of the moment arm is connected to a force transferring element in the third connection point.

The pivot axes of the first linking members mutually may define a polygon of n'th order. Similarly the pivotal axes of the second linking members may mutually define a polygon of n'th order. As an example at least one of the polygons defined may be equilateral and or defined in one plane. Further more the angles between each of the sides of at least one of the polygons may be at least substantially equal.

At least one of either the proximal or distal members may be connected to a static space truss structure or a tubular body. By attaching a static space truss or a tubular body the characteristics of the device may be changed substantially. By placing a truss or body below the device attached to the proximal member the device may be adapted to reach inside narrow spaces and thereby the device may be capable of operating in advanced and new fields e.g. of automation. By placing a truss or body above the device attached to the distal member the device the sphere of reachability is enhanced. As an example a device attached to a truss or a tubular body may be used for applications such as for animatronics, as for space robots, as for motion simulation, as a robot placer or for manipulation of large work pieces.

One advantage of using a space truss is that a high strength to weight ratio may be achieved. Especially when designing applications with large sphere of reachability compared to the size of the device it is appreciated to avoid unnecessary weight. The lower weight confers a smaller deflection caused by the weight of the device itself thereby allowing for a more precise control of the position of the distal member in relation to the proximal member. By using a space truss the elements are only influenced by tension or compression applied at the ends of the members. On the other hand a tubular body has a relatively high stiffness towards torsion around the longitudinal axis of the tube.

The power driven means may be attached to the space truss or the tubular body so as to transfer a force co-extending from the power driven means to the space truss or tubular body.

The space truss may comprise a base member and a top member facing each other and being interconnected by 2*n force transferring elements. The force transferring elements may be attached to the top member in the vicinity of n top joints and may be attached to the base member in the vicinity of n base joints. Thereby each joint connects two force transferring element to either the base member or the top member. By connecting the force transferring elements as said above the n top joints and the n base joints each defines a polygon. As an example at least one of the polygons may be equilateral. Further more the polygons each may define a polygon in one plane. Additionally the angles between each of the sides of the polygons defined by the n top joints and the n base joints respectively may be substantially equal.

The polygon defined by the pivot axes of the first linking members and polygon defined by the pivot axes of the second linking members, say first and second axes of rotation, may each define a polygon of the n'th order. In a preferred embodiment to be described in further details later, n is chosen to be three and accordingly, the polygon is a triangle.

According to a preferred embodiment of the invention, the polygon defined by the n top joints is rotated 360/(2n) degrees in relation to the polygon defined by the n base joints. Taking the example that n equals three, the polygon defining the top joints is rotated 60 degree in relation to the polygon defining the n base joints. By disposing the three (for n =3) first linking members so that the first axes of rotation defines a triangle with 60 degrees between each of the sides, it is specifically preferred to provide a device wherein the triangle defined by the first axes is rotated approximately 60 degrees in relation to the triangle (polygon of third order) defined by the n base joints. Thereby the forces necessary to raise and lower the first linking members may easily be transferred directly down to the n base joints of the space truss and thereby substantially no moments are introduced in the trusses. A device with n equalling 3 and with the triangles of the first axes rotated 60 degrees in relation to the polygon defined by the n base joints is shown e.g. in Fig. 1 and described further in the succeeding "detailed description of the invention".

According to a specific embodiment of the invention, the area of the polygon defined by the n top joints is equal to the area of the polygon defined by the n base joints.

The space truss or the tubular body may be attached to the parallel mechanism. As an example this could be done by attaching the proximal member of the parallel mechanism to the top member of either the space truss or the tubular body. Furthermore the proximal member may be constituted by the top member.

The power driven means may by constituted by at least n linear actuators exerting a force along n tension lines. The linear actuators may be hydraulic or pneumatic cylinders, electromotors or linear electrical actuators or any other kind of device capable of exerting a force along a line. As an alternative to linear actuators, or as a supplement to such actuators, the power driven means may comprise a plurality of electrically driven motors, each motor being adapted to cause rotation at one of the rotational joints. The electrically driven motors may be operatively connected to the control system.

As an example a first end of each of the n actuators may be connected to the third joint of one of the n first linking members. Thereby allowing the actuators to operate the first linking members. At the same time the n tension lines may extend through one of the n base joints. In addition the second opposite end of each of the n linear actuators may by connected to one the n base joints.

The device may be provided with a control system for controlling the power driven means, so as to control positioning and movement of the power driven means. The control system may simply be a setup of hydraulic valves for controlling the movement of hydraulic actuates. However, according to a preferred embodiment, the control system is adapted to enable robotic control over the device. I.e. to enable control of a tool-centre-point defining a position and an orientation of the distal member of the device in relation to the proximal member of the device. For the purpose of controlling the position and orientation of the distal member, the device may have an electronic information storage for storing control information related to a kinematic structure of the device. As an example the information may relate to the length of each of the elements of the device or to the angular limits of each of the rotational or spherical joints of the device. The information may further relate to the frictional resistance of each of the joints (static or dynamic). The information may further relate to the characteristics of the power driven means. If the power driven means is constituted by linear actuators, the characteristics of the actuators would naturally be the length of stroke of each of the actuators, the maximal speed of the actuators, the maximum acceleration of each of the actuators.

Preferably, the control system is allowed to access the information related to a kinematic structure of the device. Thereby the control system may be enabled to take into account the kinematic structure of the device when controlling positioning and orienting of the distal member in relation to a position and an orientation of the proximal member. The control system may as an example be connected for a period of time to one device and for another period of time to another device having a different kinematic structure. When the control system is connected to the device, the necessary kinematic information may be downloaded from the device. Accordingly, the control system will automatically adapt the necessary parameters for a specific device.

Preferably, control information stored in the electronic data storage further comprises information related to a static design value of the device. The control system may thus be adapted to access said information and thereby take into account the static design value of the device when controlling positioning and orienting of the distal member in relation to a position and an orientation of the proximal member. As an example, the control system may take deflection of the device into account when moving the tool centre point of the distal member to a specific position and orientation.

Each of the modules may further comprise means for storing information related to a static design value of each of the modules, and the control system may be adapted to access the information so as to take into account the static design value of each of the modules when controlling positioning and orienting the object in relation to the base member. The means for electronically storing information may be based on standard computer- readable storage means such as any kind of memory, e.g. RAM, ROM, EPROM, EEPROM, or any kind of disk media, including any appropriate kind of commercially available disk drive.

The control system may preferably be adapted to compute a path to be followed by a tool centre point defined on the distal member of the device. When the tool centre point is to be moved from one position in space to another, either while processing a work piece or while translating the tool centre point from one point of action to another. That path computed by the control system may, e.g., be the shortest distance, or it may be a path prescribed by certain obstacles to be avoided by the object. Similarly, the control system may be adapted to control movement of the individual modules, so that each or some of the modules may follow a prescribed path when being moved by the power driven means from one position in space to another.

Preferably the manipulator should be protected from entering singularities, i.e. joint configurations where joints are locking, whereby the mechanism jams. The second order singularities may easily lead to destruction of mechanical parts and/or destruction of the power driven means. For the device according to the present invention, singularities are likely to occur between the first and second linking members.The device accordingly may have an electronic circuit adapted to enable each of the second joints to be limited in their capability to rotate, the limitation being based on the actual position of the moving parts of the device. The electronic circuit could have a control system connected to a sensor sensing the rotational position of the second joints. Based on the sensed signals, the power driven means should be controlled so that second order singularities are avoided. The risk of second order singularities may be further reduced by designing the links between the first and second linking members in such a way, that the singular position is impossible due to a mechanical limit in the link.

According to a preferred embodiment of the invention, the distal member of the device is provided with attachment means for attaching an object to the device. As an example, the distal member may be provided with griping means for gripping an object. As an alternative, the distal member may be provided or form as a flange for mounting typical robot tools such as a welding gun, a paint gun, a power drill, a sand blasting gun gripping means for gripping work pieces or any similar kind of equipment. According to a preferred embodiment of the invention, the second spherical joints connecting the first and second linking members are constituted by three degrees of freedom, where each freedom allows rotation around one axis, i.e. in total, rotation around three axes are allowed. As an example, a second joint may be provided with a first rotational freedom between the first linking member and a hinge, and a second rotational freedom between the second linking member and the hinge, wherein

- the first rotational freedom allowing rotation around a first axis of rotation, - the second rotational freedom allowing rotation around a second axis of rotation, and

- the hinge allowing rotation around an third axis of rotation,

wherein the first and second axes of rotation each are substantially perpendicular to the third axis of rotation. The distance between each of the rotational axes may be chosen freely, e.g. based upon the size of the manipulator. As an example a traditional cardan joint or universal joint of the type known in the art, e.g. for transferring moment from an engine to the wheels of a car may be chosen.

All rotational and/or spherical joints of the device should preferably be selected with the consideration that the device should be operable in any orientation e.g. up side down or sideways. Therefore all joints should be selected with as little clearance as possible. As an example, all bearings should be of a type which allows pretension.

According to a second aspect, the present invention provides a device for movement and positioning of an object in space, said device comprising:

- power driven means for causing relative movement between the first and second linking members,

- each of the first linking members having a first rotational joint for connection of the first linking member to the proximal member and a spherical joint for connection of the first linking member to the second linking member, the first rotational joints allowing the first linking member to rotate in relation to the proximal member around a first axis of rotation and the second rotational joint allowing the first linking member to rotate in relation to the second linking member around a first point of rotation - each of the second linking members having a first rotational joint for connection of the second linking member to the distal member and a second spherical joint for connection of the second linking member to the first linking member, the first and second linking members thereby being spherically interconnected by their respective second joints,

one of either the proximal or distal members being connected to a static space truss structure.

The device according to the second aspect thus combines the features of a parallel mechanism with a static space truss attached either to the proximal or distal member by integrating a static space truss with a parallel mechanism. Alternatively a tubular body may substitute the space truss.

The device according to the second aspect of the invention may preferably be controlled similarly to the device according to the first aspect of the invention and by the use of a combination between a control system and one or more information storages for storing information related to the kinematic and static structure of the device of parts of the device.

Singularities should be handled in a similar way as described for the device according to the first aspect of the invention.

The ability of attaching an object such as a robot wrist or tool to the distal member of the device is similar to what has been described for the device according to the first aspect of the invention.

The second spherical joints of the first and second linking members may be constructed in a way similar to what has been described for the device according to the first aspect of the invention, e.g. by use of a universal joint, or by use of three rotational joints interconnected to form a spherical joint. According to a third aspect, the present invention relates to a robot comprising:

- m devices of the kind described above, m denoting the total number of devices in the robot, m>1 , the distal member of the first device being connected to the proximal member or the base member of the second device, vice versa , and

- a control system for controlling the power driven means of the m devices, so as to control positioning and orientation of the distal member of the m'th device in relation to the proximal member of the 1st device.

Each of the m devices could be interconnected, e.g. by positioning a second device on the top of the first device so that the proximal member of the second device can be bolted to the distal member of the first device and vice versa until the m'th device. For the purpose of easing the assembly of the robot, the distal member and the proximal member of each device may be provided with guiding pins and corresponding holes or similar means for guiding the proximal member of one device to a predetermined attachment position on the distal member of another device. Each of the devices may further be provided with means for easy and fast connection of the devices of the robot, e.g. by providing the proximal members and the distal members with standardised bolt holes or clamps for clamping the members together. The members may even be assembled by welding or gluing the members together or a proximal member of one device may be the distal member of another device. Similarly, one or more space trusses or even tubular bodies may be inserted between the devices so as to enhance the reachability of the robot.

The control system for controlling the movement and position of the distal member of the m'th device in relation to the proximal member of the first device may be adapted to control the robot traditionally. As an example, the system may be adapted to control each of the power driven means according to a pre-specified algorithm for controlling a robot in a Cartesian space. Alternatively, more advanced models for controlling a manipulator with a multiplicity of joints may be applied. One solution for the control of the robot may be to adapt the off-line programming system THOR™ from the Danish company AMROSE A/S. Preferably, the control system is programmed so as to control the position of a tool centre point of the distal member. In the present context, the expression "tool centre point" should be interpreted as being any suitable point or domain of the distal member serving as a reference point for the control system for controlling positioning and movement of the distal member of the last device (m'th device) in relation to the proximal member of the first device or in relation to a base member of the first device if such base member exists.

The control system may be adapted to receive information from a CAD system (Computer Aided Design system) or from a process-planning computer system. In particular, the control system may be adapted to process information related to the shaped of a work piece to be processed, e.g. by welding or painting, and to control movement of the modules of the manipulator in relation to each other in accordance with a certain path to be followed, according to the shape of the work piece to be processed.

The electronic components of the control system may be mounted on the manipulator or the modules of the manipulator, or they may be arranged separately at a location remote from the manipulator.

Preferably, the control system is adapted to compensate for any errors occurring due to the load of the devices causing some degree of deflection of the tool centre point in relation to a computed position and due to the kinematic structure of the individual devices of the robot. Thus, each of the devices may as earlier mentioned have means for storing information related to a kinematic structure of the device. Preferably the control system is adapted to access that information so as to take into account the kinematics structure of each of the devices when controlling the positioning and orienting of the distal member of the m'th device in relation to the proximal member of the first device or in relation to another reference position and orientation, e.g. of the base member of the first device, if the first device is attached to a space truss.

Each of the m devices may, as earlier mentioned further have means for electronically storing information related to a static design value of each of the devices. Preferably, the control system of the robot is adapted to access such information so as to take into account the static design value of each of the modules when controlling positioning and orienting the object in relation to the base member. According to an alternative embodiment of the invention, the control system operates by delegating control tasks to at least one of the control systems of the m devices.

The robot or the devices according to the invention may be applied for numerous applications of which examples are:

Animatronics: The superior strength of the device may be used for manipulating large constructions, e.g. for manipulation of animals in theme parks or for manipulation of simulation cabins, e.g. for training pilots or for entertainment purpose.

The device may further be used, e.g. for space robot applications where large reachability and precision is typically required.

The device could successfully be adapted for work piece manipulation, e.g. for handling large moulded or casted constructions. As an example, the device may be adapted for handling casted blocks of iron, e.g. for moving the blocks in relation to a sanding or grinding machine for deburring of the casted piece.

Moreover, the device may be adapted for welding, painting, cutting and similar mechanical working processes.

The robot according to the invention allows for automation or semi-automation of until now impossible tasks due to a highly modular concept. The invention and the applications for the device and robot should not be limited to the disclosed examples.

Detailed description of the invention

A preferred embodiment of the invention will now be described in details with reference to the drawing in which:

Fig. 1 shows a manipulator according to the present invention,

Fig. 2 shows a functional view of the manipulator Fig. 3 shows a functional view of the manipulator of Fig. 2 separated into a static and a dynamic part,

Fig. 4 shows a number of modules for a manipulator according to the invention,

Fig. 5 shows a number of combinations of modules for a manipulator,

Fig. 6 and 7 shows a model of the manipulator with defined bodies and constraints,

Fig. 8 and 9 shows different combinations of manipulators,

Fig. 10 shows a combination of a manipulator and a regular robot for painting,

Fig. 11 shows a robotic set-up for welding large steel structures,

Fig. 12 shows a flexible robot set-up mounted to a displaceable gantry,

Fig. 13 shows examples of robotic applications for processing large steel structures,

Fig. 14 shows an example of the manipulator adapted for painting of containers,

Fig. 15-23 illustrates a solution to the elimination of singularities and joint limits,

Fig. 24 shows an example of the manipulator adapted for animatronics, and

Fig. 25 shows an example of the manipulator adapted for motion simulation.

Referring to Fig. 1 a manipulator according to the present invention consists of a double- tripod mechanism 1 and a static truss 2 mounted with 3 linear actuators 3. The double tripod mechanism 1 is mounted on the static truss 2 and the linear actuators are connected between the double tripod and the remote end of the static truss 2, so as to thereby form a truss structure.

The double tripod mechanism has a proximal member 4, a distal member 5 and three pairs of linking members 6. The pairs of linking members are connected respectively to the distal member and to the proximal member by means of at least 6 hinges 7 forming 6 pivotal connection points, three between the linking members and the distal member and three between the linking members and the proximal member. The pairs of linking members each has a first 8 and a second 9 linking member interconnected by a hinge 10 forming a pivotal connection.

The first and the second linking members each have additional rotational axes 13, 14 perpendicular to the pivotal connection formed by the hinge 10. The three rotational degrees of freedom given by the hinge 10 and the rotational axes 13,14 could be substituted by a spherical joint connecting the first 8 and the second 9 linking member.

The proximal member 4 and the distal member 5 are made e.g. from steel or aluminium plates or by means of a composite material such as carbon reinforced epoxy polymer. Preferably the members have a triangular shape, and in order to optimise the weight of the members the centre of the members may be hollow, thus forming a triangular ring - as seen in Fig. 1. If the members are made from a composite material they may further have a hollow structure or they may have a kernel made e.g. from a foam material. The members could also be made from metal or plastic profiles welded or glued together. In the corners of the triangle hinge points are provided e.g. fastening adjustable hinges 7 in a reinforced part of the structure.

The linking members 8 and 9 are preferably made from bars of steel, plastic or from a composite material. The bars are connected in the corners either by means of corner connection members or simply by welding or gluing the bars together. The members thus form a frame or truss structure. In each of the sets of linking members 6, one linking member 8 has a flat frame or truss structure with a triangular shape and one linking member 9 has a three-dimensional frame or truss structure made from at least 6 bars thus forming a pyramid shape. At the 3 dimensional members power driven means can be attached to one corner point and the other corner points can be used for fastening the linking member pivotally respectively to one of the proximal or distal members - using the hinge 7 - and to the other linking member - using the hinge 10 - in that set of linking members.

The static truss 2 may have any shape and be made from any material offering the necessary strength. In order to optimise the weight of the manipulator a structure seen in Fig. 1 is preferred. The static truss is adapted to carry the tripod mechanism and to provide counter-pressure for the linear actuators 3. The linear actuators are pivotally mounted to the base of the static truss as seen in Fig. 1. The static truss may furthermore be adapted to carry power supply for the actuators and electronic equipment 12 for the control of the manipulator.

The power driven means does not have to be linear actuators but can be any means for causing a rotation of the linking members around the pivotal connections, e.g. rotational actuators mounted directly in the pivotal connections or operated to rotate the members by means of chains or belts.

The pivotal connections disclosed in the figures are of a type allowing rotation around an axis of rotation - rotational or revolute joints. The connections could just as well be of a type allowing rotation around a sphere.

As visualised in Fig. 5 and in Fig. 8 more manipulators may be connected in series if the bottom-plane of the static truss of one size manipulator holds the same size as the upper- plane of the double-tripod mechanism from another manipulator.

Manipulators of different sizes may be assembled in series e.g. on the conditions that:

- The manipulators are designed in such a way that the upper-plane of the double tripod mechanism and the bottom plane of the static truss are of different sizes, and

- the upper-plane of the double-tripod manipulator in one mechanism fits the bottom- plane of the static truss of the successor manipulator.

The manipulators may be connected in series by assembling the two double-tripods against each other or by assembling the two static trusses against each other. Manipulators with different characteristics are then created due to the asymmetric nature of the manipulators.

Two double-tripods may share the same static truss and a compact assembly is then formed. The double-tripods may be of the same size using a static truss with upper and bottom planes of the same size, or the double-tripods may have different sizes using a using a static truss with upper and bottom planes of different size. The manipulator can also be made with a number of double-tripods connected to each other and omitting the static truss. Linear actuators may in that case be mounted directly between the double-tripods or the power driven means could be adapted in form of rotational actuators mounted in the pivotal connections between the linking members.

Moreover, two tripods of different sizes may be combined in a double-tripod.

By combining the above methods for assembling manipulators a wide range of different manipulators may be build using very few different building blocks.

Motion Planning Techniques for the manipulator

In the following a technique, which enables motion planning for a very wide variety of robot mechanisms, including a manipulator according to the present invention will be described. The description is towards how to model the kinematics structure of mechanisms by constraint dynamics and how to do motion planning by including artificial forces in the constraint dynamics model.

Natural Approach

In order to model the kinematics of a manipulator,

1. Define bodies Ba, Bp1a, Bp2a, Bp3a, Bp1b, Bp2b, Bp3b, and Bb. Arrange these bodies in a "legal/physical" initial configuration of the manipulator - see Fig. 6. 2. Constrain Bp1a, Bp2a, and Bp3a to Ba by 3 sets of 5 constraints defining the proper revolute joints, as indicated in Fig. 6. 3. Constrain Bp1b, Bp2b, and Bp3b to Bb in the same manner. Having a "legal/physical" initial configuration of the mechanism, p1a and p1b will be coincident - the same applies for p2a/p2b and p3a/p3b. 4. Now select the point a12 on a line normal to the plane defined by the three points a1 , a2, and p2a. The line must intersect the line segment connecting a1 and a2, dividing the segment in two pieces of equal length. The distance between a12 and the plane must equal the distance between a1 and a2. Define points a13, a23, b12, b13, and b23 in a similar manner. 5. Constrain a1 , a2, and a12 to p2b by demands of fixed distances. Fix the distances between a2/a3/a23 and p3b in the same way. Finally, constrain a3/a1/a13 to p1 b.

In total, 8 bodies with 8 x 6 = 48 degrees of freedom (DOF) ail together are defined. Each constraint removes one of these DOF. 6 revolute joints with 6 x 5 = 30 constraints and further 3 x 3 = 9 constraints are introduced. Thus, there remains 48 - 30 - 9 = 9 DOF of which 3 DOF are the internal DOF of the VGT mechanism.

The bodies defined relate closely to the actual bodies in the physical mechanism, and the application of the various types of artificial forces is straightforward.

Efficient Approach

An alternative approach is the following:

1. Define bodies Ba, and Bb, and point masses p1 , p2, and p3 as indicated in Fig. 7. 2. Using constraints of fixed distance, constrain a1 , a3, b1 , and b3 to p1.

3. Similarly, constrain a1 , a2, b1 , and b2 to p2.

4. Finally, constrain a2, a3, b2, and b3 to p3.

Two bodies and 3 point masses with a total of 2 x 6 + 3 x 3 = 21 DOF are defined. In total, 3 x 4 = 12 constraints are introduced. The remaining number of DOF again is 21 - 12 = 9 DOF, of which 3 DOF are the internal DOF of the manipulator.

The computation of artificial forces, e.g. repulsion, acting on the bodies defined above require special treatment in addition to the technique described by Lars Overgaard: Applied Robot Motion Planning using Artificial Potentials and Distributed Control, Ph.D. thesis, The Maersk Mc-Kinney Moller Institute for Production Technology, Odense

University, Denmark). For example, an obstacle in the region between a1 , a2, and p2 will be quite close to the physical manipulator, but far from both Ba and p2. Thus, the computation of minimum distance points between the manipulator and obstacles in the environment must use a separate set of bodies representing the manipulator, like the one described in the section Natural Approach. The efficient model defined here continuously updates the configuration of this natural model. The artificial repulsion computed on basis of the natural model is distributed on the relevant bodies in the efficient model, e.g. Ba and p2, using an appropriate distribution function. The closer the point of force excertion is to one body relative to other involved bodies, the larger a fraction of the total force is excerted on this particular body. Actuators

During the simulation/motion planning process, the natural co-ordinates for the mechanism are the co-ordinates of the individual bodies. Positions, velocities, and accelerations of the actuators may be computed at any instant from the actual positions, velocities, and accelerations of the bodies.

Modularity

When motion planning for a complex mechanism including - among other things - several manipulators, either of the above approaches may be used to model each manipulator. Any number of modules can be incorporated in the model.

Modular Control

More manipulators combined into a complex mechanism may preferably be controlled modulariy. By adapting control units in each manipulator and interconnecting the control units to form one overall control unit a highly modular robotic concept can be achieved.

Modular control is a well-known paradigm in industrial applications. The typical solutions are based on a number of small and fairly simple programmable units handling I/O signals and simple low level control. The units are connected by means of industrial network technologies, which can be connected to calculation units such as industrial computers and to user interfaces such as hand held terminals etc.

The existing products and standards are primarily intended for industrial automation wherein the system architecture remains over a long period of time and wherein the hardware and software related structure therefore is static. The static structure of the present industrial robot systems affects the applicability of the robots e.g. for short-term application, low batch sizes etc. in that the connectivity between different robots and peripheral systems is expensive and time-consuming.

Each manipulator that is interconnected into a robotic system could e.g. be equipped with a dedicated control system, designed and programmed to handle the control of that specific manipulator. The control modules adapted in the individual manipulator carry all relevant information about the geometry, dynamic and kinematics structure relevant for the control of that manipulator. By means of a generic interface the manipulator provides the information - along with information on how to control that manipulator, to a network of interconnected manipulators. A master unit collects information from all interconnected manipulators and controls the manipulators as if they were one robot. If one manipulator is replaced the master unit can automatically update the overall robot structure and thus the control algorithms therefore.

Each of the manipulators can also communicate independently of the master unit, and e.g. send information about deflection, load etc. This information enables each manipulator to compensate for inaccuracies. As an example, the first manipulator in a row of interconnected manipulators sends a signal that it can not by itself compensate for a certain deflection. The succeeding manipulators in the row can then co-operate in order to remove the error.

Each manipulator can likewise be connected to sensors that may be used in connection with the control of previous or succeeding manipulators in a row.

Finally the motion planning tasks for the row of manipulators can be calculated distributed in each of the interconnected manipulators. In that case a master unit may not be needed at all.

Fig 10 shows a robotic snake composed of a number of manipulators according to the present invention and a standard type industrial 7-axis robot. The working range and flexibility of the snake is far better than what have been seen using traditional robots with a traditional kinematics structure.

Fig. 11 -14 shows different highly flexible robot set-ups using the manipulator and a regular "anthropomorphic" type robot to perform various tasks such as painting, welding, sandblasting and quality inspecting on large work pieces. The applications shown could be used e.g. in heavy industries for shipbuilding, bridge construction or for assembling large steel constructions in tall buildings.

The first and the second linking members each have additional rotational axes 13, 14 perpendicular to the pivotal connection formed by the hinge 10. The three rotational Joint singularities and Workspace

The workspace of parallel mechanisms preferable has to be designed so that no singularities occur, i.e. joint configurations where the mechanism jams and possible damage occurs. Referring to Fig. 15-17, such a singular configuration appears if the rotational members 15,16 of the first and the second linking members 8 and 9 are brought to a position where the axes 13,14 coincides, and the members 15,16 will then be able to rotate unconstrained about their coinciding axes 13,14. Any minor such unconstrained rotation of members will result in jamming of the first and second linking members 8,9 and the actuator 19 connected to the linking member 16 is no longer able to move the linking member 9. If this situation occurs the actuator force may damage the mechanism.

The mechanism may be designed having no joint singularities in the workspace by mechanically restricting the stroke of the actuators in the worst case configuration The worst case situation occurs when all three actuators are mutually elongated as shown in Fig. 15. Fig. 15 shows a situation wherein all three linking members are close to singularity.

This traditional method of eliminating the singularities is very safe, but the workspace is rather limited, as the joint between the first and second linking members 8,9 is far from singularity when only one actuator is elongated, this is shown in Fig. 16. In Fig. 16 the actuator 19 is in the same position as in Fig. 15, but the other two actuators 20, 21 are retracted. The members 15 and 16 are now far from the singular position and the belonging actuator 19 may be further elongated in order to obtain a larger workspace.

The additional impact on the workspace caused by this additional stroke is illustrated in Fig. 17.

Joint limits and workspace The same considerations are valid when focussing on the critical joint limit occurring when all actuators are retracted. In Fig. 18, all 3 joints 22, 23 and 24 (24 is not shown in the Fig.) are close to a limit, but as illustrated in Fig. 19 the joint 25 is far from a joint limit, even though the actuator 26 is in the same position as indicated for the actuator 27 in Fig. 18. The actuator to the right in Fig. 19 may be even further retracted and the workspace may thereby be further increased. This is shown in Fig. 20. Control strategies for accomplishing an optimised workspace

The stroke limits of a directly controlled actuator must depend on the position of the other two actuators in order to as well optimise the workspace as to avoid joint singularities and joint limits of the indirectly controlled joints. The stroke limits of the actuators must thus be software controlled and not governed by mechanical limits in the actuators.

This way of controlling the mechanism may impose larger risks for damaging the mechanism especially during commissioning of new mechanisms or when the position measuring devices are out of calibration.

This risk can be reduced by incorporating a sensor in the indirectly controlled joint between the members 15 and 16 (Fig. 15) giving signal to the control system or to the safety system, when the joint is to close to a singularity or a limit.

In severe situations, where the control system is malfunctioning, for instance during commissioning, this way of securing the mechanism may not be adequate.

The risk may be further reduced by designing the link between members 15 and 16 with reference to Fig. 15, in such a way that the singular position is impossible due to a mechanical limit in the link. This mechanical limit must be able to support the torque in the joint in a safe manner.

Figs. 21-23 illustrate 3 different ways to realise such a mechanical limit in the joint.

Analogous the link may be designed in such a way that when the joint limit is reached, the contact forces are supported in a safe manner.

Referring to Fig. 24, the manipulator may be adapted for animatronics. The head 28 of the dinosaur 29 is moved by a manipulator 30.The manipulator is controlled by a computer system capable of animating the motion of the dinosaur and to plan the motion of the actuators 31 accordingly. The manipulator provides an excellent stability to weight ratio for the application.

Fig. 25 shows an example of the manipulator adapted for motion simulation. The manipulator 32 offers the sufficient strength, stability and speed to move the cabin 33 in relation to an animation visualised on the screen 34 in front of the user. Moreover the size and weight of the manipulator enables a portable setup of the cabin and the manipulator.

Claims

1. A device for movement and positioning of an object in space, said device comprising:

the first linking members comprising:

- at least two rigid force transferring elements, each of the elements at a first end being mutually interconnected in a connection point and at a second opposite end being connected to the first linking member so as to define a truss, and

2. A device according to claim 1 , wherein the second ends of the at least two rigid force transferring elements are connected to the first linking members in the vicinity of the first rotational joints and in the vicinity of the second spherical joints of the first linking members.

3. A device according to claim 1 or 2, wherein the force transferring elements are mutually interconnected to form a space truss structure in such a way that any force vector implied by the power driven means to the force transferring elements is converted into force components extending substantially through the first, second and third joints, respectively.

4. A device according to any of the preceding claims, wherein the first linking members define space truss structures each with at least 6 force transferring elements connected in at least 4 connection points, the first joints of the first linking members allowing rotation around pivot axes parallel to lines extending between a first and a second of the 4 connection points, the second joints of the first linking members being disposed in the vicinity of a third of the 4 connection points and the third joints of the first linking members being disposed in the vicinity of a fourth of the 4 connection points.

5. A device according to claim 4, wherein the space truss structure of the first linking members defines a tetrahedron.

6. A device according to claim 4 or 5, wherein the pivot axes of the first linking members mutually defines polygon of n'th order.

7. A device according to claim 6, wherein the defined polygon is equilateral.

8. A device according to claim 6 or 7, wherein the first linking members defines the polygon in one plane.

9. A device according to any of claims 6 - 8, wherein the angles between each of the sides of the polygon are at least substantially equal.

10. A device according any of claim 1 - 9, wherein at least one of either the proximal or distal members is connected to a static space truss structure and wherein the power driven means is transferring compressive strength from the space truss to the third joint.

11. A device according to claim 10, wherein the static space truss comprises a base member and a top member facing each other and being interconnected by 2*n force transferring elements, the force transferring elements being attached to the top member in the vicinity of n top joints and to the base member in the vicinity of n base joints, each joint thereby connecting two force transferring element to either the base member or the top member.

12. A device according to claim 11 , wherein the polygon defined by the n top joints is equilateral.

13. A device according to claim 11 or 12, wherein the n top joints defines a polygon in one plane.

5 14. A device according to any of claims 11-13, wherein the angles between each of the sides of the polygon defined by the n top joints are at least substantially equal.

15. A device according to any of claims 11-14, wherein the polygon defined by the n base joints is equilateral.

10

16. A device according to any of claims 11-15, wherein the n base joints defines a polygon in one plane.

17. A device according to any of claims 11-16, wherein the angles between each of the 15 sides of the polygon defined by the n base joints are at least substantially equal.

18. A device according to any of claims 11-17, wherein the polygon defined by the n top joints is rotated 360/(2n) degrees in relation to the polygon defined by the n base joints.

20 19. A device according to any of claims 11-18, wherein in the area of the polygon defined by the n top joints is equal to the area of the polygon defined by the n base joints.

20. A device according to any of claims 11-19, wherein the proximal member is attached to the top member.

25

21. A device according to any of claims 11-20, wherein the proximal member is constituted by the top member.

22. A device according to any of claims 1-21, wherein the power driven means is 30 constituted by at least n linear actuators exerting a force along n tension lines.

23. A device according to claim 22, wherein a first end of each of the n actuators are connected to the third joint of one of the n first linking members and wherein the n tension lines extends through one of the n base joints.

35

24. A device according to claim 22 or 23, wherein the second opposite end of each of the n linear actuators are connected to one the n base joints.

25. A device according to any of the preceding claims, wherein n equals three. 5

26. A device according to any of the preceding claims, further comprising a control system for controlling the power driven means, so as to control positioning and movement of the power driven means.

10 27. A device according to claim 26, wherein the control system is adapted to control positioning and orientation of the distal member in relation to a position and an orientation of the proximal member.

28. A device according to any of the preceding claims, further comprising an electronic 15 information storage for storing control information related to a kinematic structure of the device,

29. A device according to claim 28, wherein the control system is adapted to access the information related to a kinematic structure of the device, and wherein the control system

20 is adapted to take into account the kinematics structure of the device when controlling positioning and orienting of the distal member in relation to a position and an orientation of the proximal member.

30. A device according to claim 28 or 29, wherein the control information further

25 comprises information related to a static design value of the device, and wherein the control system is adapted to access said information so as to take into account the static design value of the device when controlling positioning and orienting of the distal member in relation to a position and an orientation of the proximal member.

30 31. A device according to any of the preceding claims, being protected from entering singularities.

32. A device according to claim 31 , wherein the protection comprises an electronic circuit adapted to enable each of the second joints to be limited in their capability to rotate, the limitation being based on the position and orientation of the distal member in relation to the proximal member.

33. A device according to claim 32, wherein the electronic circuit comprises a control 5 system connected to at least one sensor adapted to indicate the rotational position of the second joints, and wherein the control system is adapted to receive signals from the sensor and based on the received signals, to control the power driven means so that second order singularities are avoided.

10 34. A device according to any of claims 31-33, wherein the protection further comprises mechanical limitations of the second joints.

35. A device according to any of the preceding claims, wherein the distal member comprises attachment means for attaching an object to the device.

15

36. A device according to any of the preceding claims, wherein the second spherical joints connecting the first and second linking members comprises a first rotational freedom between the first linking member and a hinge, and a second rotational freedom between the second linking member and the hinge, wherein

20

- the first rotational freedom allowing rotation around a first axis of rotation,

- the second rotational freedom allowing rotation around a second axis of rotation, and

- the hinge allowing rotation around an third axis of rotation,

25 wherein the first and second axes of rotation each are substantially perpendicular to the third axis of rotation.

37. A device according to any of the preceding claims, wherein the second spherical joints 30 connecting the first and second linking members comprises a cardan.

38. A device for movement and positioning of an object in space, said device comprising: - a proximal member and a distal member facing each other and being interconnected by n>2 linking members, each linking member comprising a first and a second linking member,

- each of the first linking members having a first rotational joint for connection of the first linking member to the proximal member and a spherical joint for connection of the first linking member to the second linking member, the first rotational joints allowing the first linking member to rotate in relation to the proximal member around a first axis of rotation and the second spherical joint allowing the first linking member to rotate in relation to the second linking member around a first point of rotation

39. A device according to claim 38, wherein the static space truss comprises a base member and a top member facing each other and being interconnected by 2*n force transferring elements, the force transferring elements being attached to the top member in the vicinity of n top joints and to the base member in the vicinity of n base joints, each joint thereby connecting two force transferring element to either the base member or the top member.

40. A device according to claim 39, wherein the polygon defined by the n top joints is equilateral.

41. A device according to claim 39 or 40, wherein the n top joints defines a polygon in one plane.

42. A device according to any of claims 39-41 , wherein the angles between each of the sides of the polygon defined by the n top joints are at least substantially equal.

43. A device according to any of claims 39-42, wherein the polygon defined by the n base 5 joints is equilateral.

44. A device according to any of claims 39-43, wherein the n base joints defines a polygon in one plane.

10 45. A device according to any of claims 39-44, wherein the angles between each of the sides of the polygon defined by the n base joints are at least substantially equal.

46. A device according to any of claims 39-45, wherein the polygon defined by the n top joints is rotated 360/(2n) degrees in relation to the polygon defined by the n base joints.

15

47. A device according to any of claims 39-46, wherein in the area of the polygon defined by the n top joints is equal to the area of the polygon defined by the n base joints.

48. A device according to any of claims 39-47, wherein the proximal member is attached 20 to the top member.

49. A device according to any of claims 39-47, wherein the proximal member is constituted by the top member.

25 50. A device according to any of claims 39-49, wherein the power driven means is constituted by at least n linear actuators exerting a force along n tension lines.

51. A device according to any of claims 50, wherein the first linking members comprise a third joint for connection of the linking members to the actuators, the third joint being

30 disposed offset from a plane extending the first axes of rotation and the first point of rotation.

52. A device according to claim 51 , wherein a first end of each of the n actuators are connected to the third joint of one of the n first linking members and wherein the n tension

35 lines extends through one of the n base joints.

53. A device according to claim 50-52, wherein the second opposite end of each of the n linear actuators are connected to one the n base joints.

5 54. A device according to any of claims 38-53, wherein n equals three.

55. A device according to claims 38-54, further comprising a control system for controlling the power driven means, so as to control positioning and movement of the power driven means.

10

56. A device according to claim 55, wherein the control system is adapted to control positioning and orientation of the distal member in relation to a position and an orientation of the proximal member.

15 57. A device according to any claims 38-56, further comprising an electronic information storage for storing control information related to a kinematic structure of the device,

58. A device according to claim 57, wherein the control system is adapted to access the information related to a kinematics structure of the device, and wherein the control system

59. A device according to claim 57 or 58, wherein the control information further

30 60. A device according to any of claims 38-59, being protected from entering singularities.

61. A device according to claim 60, wherein the protection comprises an electronic circuit adapted to enable each of the second joints to be limited in their capability to rotate, the limitation being based on the position and orientation of the distal member in relation to 35 the proximal member.

62. A device according to claim 61 , wherein the electronic circuit comprises a control system connected to at least one sensor adapted to indicate the rotational position of the second joints, and wherein the control system is adapted to receive signals from the

5 sensor and based on the received signals, to control the power driven means so that second order singularities are avoided.

63. A device according to any of claims 60-62, wherein the protection further comprises mechanical limitations of the second joints.

10

64. A device according to any of claims 38-63, wherein the distal member comprises attachment means for attaching an object to the device.

65. A device according to any of claims 38-64, wherein the second spherical joints 15 connecting the first and second linking members comprises a first rotational freedom between the first linking member and a hinge, and a second rotational freedom between the second linking member and the hinge, wherein

20 - the second rotational freedom allowing rotation around a second axis of rotation, and

- the hinge allowing rotation around an third axis of rotation,

wherein the first and second axes of rotation each are substantially perpendicular to the 25 third axis of rotation.

66. A device according to any of claims 38-65, wherein the second spherical joints connecting the first and second linking members comprises a cardan.

30 67. A robot comprising:

- m devices according to any of the claims 1-37 or 38-66, m denoting the total number of devices in the robot, m>1 , the distal member of the first device being connected to the proximal member or the base member of the second device, vice

35 versa , and - a control system for controlling the power driven means of the m devices, so as to control positioning and orientation of the distal member of the m'th device in relation to the proximal member of the 1st device.

5

68. A robot according to claim 67, wherein the control system is adapted to access the information related to a kinematic structure of the m devices by accessing the electronic information storages of each of the m devices, and wherein the control system is adapted to take into account the kinematic structure of each of the m devices when controlling the

10 positioning and orientating of the distal member of the m'th device in relation to the proximal member of the first device.

69. A robot according to claim 67 or 68, wherein the control system is adapted to access the information related to a static design value of the m devices by accessing the

15 electronic information storages of each of the m devices, and wherein the control system is adapted to take into account the static design values of each of the m devices when controlling the positioning and orientating of the distal member of the m'th device in relation to the proximal member of the first device.

20 70. A robot according to claim 67-69, wherein the control system operates by delegating control tasks to at least one of the control systems of the m devices.

71. A system for designing a modular robot, said system comprising

25 - q>1 devices according to any of claims 1-37 or according to any of claims 38-66,

- attachment means for attaching one device to another device, and

- a control system for controlling the movement and positioning of the devices, the control system being adapted for the definition of a tool-centre point in relation to one part of the robot and a reference frame in relation to another part of the robot

30 an to position and orient the tool frame in relation to the reference frame.

72. The use of a device, robot or system according to any of the preceding claims for animatronics.

73. The use of a device, robot or system according to any of claims 1-70 for motion simulation.

74. The use of a device, robot or system according to any of claims 1-70 for space 5 robotics.

75. The use of a device, robot or system according to any of claims 1-70 for work piece manipulation.

10 76. The use of a device, robot or system according to any of claims 1-70 for welding.

77. The use of a device, robot or system according to any of claims 1-70 for painting.

78. The use of a device, robot or system according to any of claims 1-70 for cutting. 15